Spurious Strange Correlators in Symmetry-Protected Topological Phases

This paper identifies and classifies three mechanisms—high-dimensional irreducible representations, symmetry phase mismatches, and symmetry-breaking long-range order—that cause ill-chosen reference states to generate spurious long-range strange correlators in trivial phases, thereby providing guidelines to avoid false positives in diagnosing symmetry-protected topological order.

The Gist

The Big Picture: The "Lie Detector" Problem

Imagine you are a detective trying to find a hidden treasure (a Symmetry-Protected Topological or SPT phase). In the world of quantum physics, these are special states of matter that look boring on the surface but have a secret, complex structure underneath.

To find this treasure, physicists use a tool called a Strange Correlator. Think of this tool as a "lie detector" or a "compatibility test."

• The Test: You take the mysterious material you are studying (the Target State) and compare it against a known, boring, simple material (the Reference State).
• The Rule: If the two materials "talk" to each other over long distances (showing a long-range correlation), the test says, "Aha! The target is a special SPT phase with a secret structure!" If they stop talking quickly, the test says, "It's just a boring, trivial material."

The Problem: The authors of this paper discovered that this lie detector can be tricked. Sometimes, a boring, trivial material can fool the test and scream "I'm special!" just because you picked the wrong reference material to compare it against. These are called Spurious Strange Correlators (fake signals).

---

The Core Discovery: Why the Lie Detector Fails

The authors used a mathematical framework called Matrix Product States (MPS) to figure out why the detector fails. They found that the test relies on a specific mathematical property called Magnitude-Degeneracy.

The Analogy: The Echo Chamber
Imagine the Strange Correlator is like shouting into a canyon and listening for an echo.

• Real SPT Phase: The canyon has a special shape (due to its secret structure) that always creates a perfect, long-lasting echo, no matter how you shout.
• Trivial Phase (The Fake): Usually, a boring canyon just absorbs the sound. But, the authors found that if you shout from a specific spot or with a specific tone (a bad choice of Reference State), even a boring canyon can create a fake, long-lasting echo.

The paper proves that this "fake echo" happens when the mathematical "transfer matrix" (the machine doing the calculation) has multiple "loudest notes" (eigenvalues) that are equally strong. When this happens, the signal doesn't die out, even if the material is boring.

---

The Three Ways to Get a Fake Signal

The authors identified three specific ways a boring material can trick the detector. Here are the three mechanisms:

1. The "Big Orchestra" Mistake (High-Dimensional Representations)

• The Scenario: Imagine your material is a simple, boring room. But, you decide to test it using a reference state that is a massive, complex orchestra (a high-dimensional representation).
• The Glitch: Even though the room is boring, the sheer complexity of the orchestra creates a mathematical "resonance" that looks like a long-range signal.
• The Paper's Example: They looked at a Spin-2 AKLT model. This is a material that is mathematically trivial (boring), but because it involves complex symmetries (SO(3)), a standard test can mistake it for a special phase.
• The Fix: You need to pick a reference state that is simple enough (a "soloist") so it doesn't create this accidental resonance.

2. The "Wrong Tune" Mistake (Phase Mismatch)

• The Scenario: Imagine you and your friend are trying to sing a duet. You are singing in a major key, but your friend (the reference state) is singing in a minor key. Even though you are both singing the same song, the clash creates a weird, lingering dissonance.
• The Glitch: If the "symmetry" of your target material and the reference material don't match perfectly (specifically, if they have different "phases" or signs under symmetry operations), the math creates a fake long-range signal.
• The Paper's Example: They showed that if you take a trivial material with a simple "flip" symmetry (like flipping a coin) and compare it to a reference state that flips the opposite way, the test will falsely say the material is special.
• The Fix: Make sure your reference state sings in the exact same "key" (symmetry representation) as the target.

3. The "Broken Mirror" Mistake (Symmetry Breaking)

• The Scenario: Imagine a room where everyone is standing still (a symmetric state). But, the room is actually in a state where people are supposed to be moving left or right (symmetry breaking), and you are looking at a weird mix of both.
• The Glitch: If the material has "broken symmetry" (like a magnet that is already pointing North), it naturally has long-range order. If you compare this to a reference state that is also symmetric, the math gets confused and sees a long-range signal, even though the signal comes from the "broken" nature of the material, not a topological secret.
• The Paper's Example: They used a GHZ state (a specific entangled state often used in quantum computing) which is not a topological phase but is highly entangled. The test picked up its long-range order and called it an SPT phase.
• The Fix: Ensure your reference state preserves the full symmetry of the system so you aren't measuring the "broken" order.

---

The Solution: How to Avoid the Trap

The paper doesn't just point out the problem; it gives a recipe for a "safe" test. To correctly identify a topological phase without getting a fake signal, your Reference State must be:

1. Trivial: It must be a simple, boring material.
2. Symmetric: It must respect all the same rules (symmetries) as the target material.
3. Matching: It must sing the exact same "tune" (1D symmetry representation) as the target.
4. Simple: It must avoid complex "orchestras" (high-dimensional representations) that cause accidental resonances.

The "Inverse Scanning" Strategy

For scientists who don't have the perfect reference state ready, the authors suggest a strategy called "Inverse Scanning."

• The Idea: Don't just test the material once. Test it against many different reference states.
• The Logic:
  • If the material is truly special (SPT), it will show a long-range signal no matter which reference state you use (because its secret structure is robust).
  • If the material is boring (Trivial), the long-range signal will disappear if you pick the right reference state. If the signal is "fragile" and vanishes with a small change in the reference, it was a fake.

Summary

This paper is a warning label for physicists. It says: "The Strange Correlator is a powerful tool, but it is easily fooled. If you pick the wrong reference state, you might think you found a new topological phase when you actually just found a mathematical glitch. To get the right answer, you must carefully choose a reference state that matches the target's symmetry and simplicity."

Technical Summary

1. Problem Statement

The strange correlator is a widely used diagnostic tool for detecting Symmetry-Protected Topological (SPT) phases. It is defined as a correlation function between a target state $|\Phi\rangle$ and a chosen trivial reference state $|\Omega\rangle$:
$$C(i, j) = \frac{\langle \Omega | O_i^a O_j^b | \Phi \rangle}{\langle \Omega | \Phi \rangle}$$
The standard criterion posits that if $|\Phi\rangle$ is a nontrivial SPT phase, the strange correlator exhibits long-range behavior (power-law decay or saturation), whereas for a trivial phase, it decays exponentially.

The Core Issue: The paper identifies a critical flaw in this methodology: the result of the strange correlator depends heavily on the choice of the reference state $|\Omega\rangle$. The authors demonstrate that an ill-chosen reference state can induce spurious long-range strange correlators in topologically trivial phases, leading to false positives in SPT diagnosis. This ambiguity has not been systematically addressed, leaving researchers without clear guidelines on how to select a valid reference state.

2. Methodology

The authors employ a rigorous analytical framework based on Matrix Product States (MPS) to study 1D gapped bosonic/spin systems.

• MPS Formalism: They express the target state $|\Phi\rangle$ and the reference state $|\Omega\rangle$ (assumed to be a product state $|\omega\rangle^{\otimes L}$) using MPS tensors.
• Transfer Matrix Analysis: The calculation of the strange correlator is reduced to the spectral properties of a specific transfer matrix $M_\omega$, constructed from the MPS tensors of the target state and the reference state vector.
• Theoretical Proofs: The authors prove theorems linking the long-range behavior of the correlator directly to the magnitude-degeneracy of the largest eigenvalue of the transfer matrix.
• Case Studies: They systematically construct counterexamples using specific lattice models (e.g., spin-2 AKLT, Z2 models, GHZ states) to illustrate distinct mechanisms causing these spurious signals.

3. Key Contributions and Theoretical Framework

A. The Origin of Long-Range Correlators: Magnitude-Degeneracy

The central theoretical finding is that long-range strange correlators arise if and only if the largest eigenvalue of the transfer matrix $M_\omega$ is magnitude-degenerate.

• Definition: An eigenvalue $\lambda_0$ is $D$-fold magnitude-degenerate if there exist $D-1$ other eigenvalues $\lambda_1, \dots, \lambda_{D-1}$ such that $|\lambda_0| = |\lambda_1| = \dots = |\lambda_{D-1}| > |\lambda_D|$.
• Theorem: If $D=1$ (non-degenerate), the connected strange correlator decays exponentially to zero. If $D>1$, there exist operators such that the correlator exhibits long-range behavior.

B. Classification of Spurious Mechanisms

The paper identifies three distinct mechanisms that cause magnitude-degeneracy in trivial phases, leading to false positives:

1. Mechanism 1: High-Dimensional Irreducible Representations (Irreps)

   • Cause: The target state (even if trivial) possesses a virtual bond space carrying a high-dimensional linear irrep of the symmetry group (e.g., the spin-2 AKLT model with SO(3) symmetry).
   • Effect: The transfer matrix inherits this degeneracy, mimicking the behavior of nontrivial SPTs (which rely on projective representations).
   • Solution: A specific "signal-to-noise" ratio optimization is proposed to select a reference state that suppresses the high-dimensional sector in favor of the 1D trivial sector.

2. Mechanism 2: Phase Mismatch in Symmetry Representations

   • Cause: A mismatch between the 1D linear representation (charge/parity) of the target state and the reference state under the symmetry group $G$.
   • Effect: Even with 1D irreps, if the target and reference transform with different phases (e.g., different eigenvalues under a spin-flip operation), the transfer matrix becomes magnitude-degenerate.
   • Solution: The reference state must transform under the same 1D symmetry representation as the target state.

3. Mechanism 3: Symmetry Breaking

   • Cause: The target state is a symmetric superposition of symmetry-broken ground states (e.g., a GHZ state in the Ising model).
   • Effect: The MPS tensor becomes block-diagonal. If the reference state is symmetric, the transfer matrix retains block-diagonal structure with degenerate leading eigenvalues across blocks.
   • Solution: The reference state must preserve the full global symmetry to distinguish SPT order from symmetry-breaking long-range order.

4. Key Results and Examples

• Spin-2 AKLT Model (Mechanism 1): The ground state is a trivial SPT (group cohomology class is trivial) but carries a 3D SO(3) irrep. Using a standard reference state, the strange correlator shows long-range behavior, falsely suggesting nontrivial topology.
• Z2 Phase Mismatch (Mechanism 2): For a trivial product state with Z2 symmetry, choosing a reference state with a different parity eigenvalue (phase mismatch) yields a constant long-range correlator, despite the absence of SPT order.
• GHZ State (Mechanism 3): A symmetric superposition of ferromagnetic states (trivial but highly entangled) yields a long-range strange correlator if the reference state is chosen to be symmetric, mimicking SPT signatures.

5. Significance and Guidelines

This work provides a rigorous "caveat" for the application of strange correlators, transforming them from a heuristic tool into a more reliable diagnostic method.

Operational Guidelines for Avoiding False Positives:
To correctly identify a nontrivial SPT phase using strange correlators, the trivial reference state $|\Omega\rangle$ must satisfy:

1. Symmetry Preservation: It must preserve the full global symmetry of the system (to avoid Mechanism 3).
2. Representation Matching: It must transform under the same 1D linear representation (same charge/parity) as the target state (to avoid Mechanism 2).
3. Spectral Gap Control: In cases involving high-dimensional irreps (Mechanism 1), the reference state must be chosen (potentially via coarse-graining or optimization algorithms) to ensure the transfer matrix's leading eigenvalue corresponds to the 1D trivial sector, not the high-dimensional noise sector.

Future Outlook:
The authors suggest an "inverse scanning" strategy for numerical studies (like QMC or ED): test the target state against a family of symmetric reference states. If the long-range correlation is robust across all valid references, it indicates genuine SPT order. If the signal is fragile (disappears with a specific reference choice), the state is likely trivial.

This paper fundamentally clarifies the conditions under which strange correlators are valid, preventing misinterpretation of trivial entanglement structures as topological order.